The following patch renames __in_dev_get() to __in_dev_get_rtnl() and
introduces __in_dev_get_rcu() to cover the second case.
1) RCU with refcnt should use in_dev_get().
2) RCU without refcnt should use __in_dev_get_rcu().
3) All others must hold RTNL and use __in_dev_get_rtnl().
There is one exception in net/ipv4/route.c which is in fact a pre-existing
race condition. I've marked it as such so that we remember to fix it.
This patch is based on suggestions and prior work by Suzanne Wood and
Paul McKenney.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Meelis Roos <mroos@linux.ee> wrote:
> RK> My firewall setup relies on proxyarp working. However, with 2.6.14-rc3,
> RK> it appears to be completely broken. The firewall is 212.18.232.186,
>
> Same here with some kernel between 14-rc2 and 14-rc3 - no reposnse to
> ARP on a proxyarp gateway. Sorry, no exact revison and no more debugging
> yet since it'a a production gateway.
The breakage is caused by the change to use the CB area for flagging
whether a packet has been queued due to proxy_delay. This area gets
cleared every time arp_rcv gets called. Unfortunately packets delayed
due to proxy_delay also go through arp_rcv when they are reprocessed.
In fact, I can't think of a reason why delayed proxy packets should go
through netfilter again at all. So the easiest solution is to bypass
that and go straight to arp_process.
This is essentially what would've happened before netfilter support
was added to ARP.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Arnaldo and I agreed it could be applied now, because I have other
pending patches depending on this one (Thank you Arnaldo)
(The other important patch moves skc_refcnt in a separate cache line,
so that the SMP/NUMA performance doesnt suffer from cache line ping pongs)
1) First some performance data :
--------------------------------
tcp_v4_rcv() wastes a *lot* of time in __inet_lookup_established()
The most time critical code is :
sk_for_each(sk, node, &head->chain) {
if (INET_MATCH(sk, acookie, saddr, daddr, ports, dif))
goto hit; /* You sunk my battleship! */
}
The sk_for_each() does use prefetch() hints but only the begining of
"struct sock" is prefetched.
As INET_MATCH first comparison uses inet_sk(__sk)->daddr, wich is far
away from the begining of "struct sock", it has to bring into CPU
cache cold cache line. Each iteration has to use at least 2 cache
lines.
This can be problematic if some chains are very long.
2) The goal
-----------
The idea I had is to change things so that INET_MATCH() may return
FALSE in 99% of cases only using the data already in the CPU cache,
using one cache line per iteration.
3) Description of the patch
---------------------------
Adds a new 'unsigned int skc_hash' field in 'struct sock_common',
filling a 32 bits hole on 64 bits platform.
struct sock_common {
unsigned short skc_family;
volatile unsigned char skc_state;
unsigned char skc_reuse;
int skc_bound_dev_if;
struct hlist_node skc_node;
struct hlist_node skc_bind_node;
atomic_t skc_refcnt;
+ unsigned int skc_hash;
struct proto *skc_prot;
};
Store in this 32 bits field the full hash, not masked by (ehash_size -
1) Using this full hash as the first comparison done in INET_MATCH
permits us immediatly skip the element without touching a second cache
line in case of a miss.
Suppress the sk_hashent/tw_hashent fields since skc_hash (aliased to
sk_hash and tw_hash) already contains the slot number if we mask with
(ehash_size - 1)
File include/net/inet_hashtables.h
64 bits platforms :
#define INET_MATCH(__sk, __hash, __cookie, __saddr, __daddr, __ports, __dif)\
(((__sk)->sk_hash == (__hash))
((*((__u64 *)&(inet_sk(__sk)->daddr)))== (__cookie)) && \
((*((__u32 *)&(inet_sk(__sk)->dport))) == (__ports)) && \
(!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif))))
32bits platforms:
#define TCP_IPV4_MATCH(__sk, __hash, __cookie, __saddr, __daddr, __ports, __dif)\
(((__sk)->sk_hash == (__hash)) && \
(inet_sk(__sk)->daddr == (__saddr)) && \
(inet_sk(__sk)->rcv_saddr == (__daddr)) && \
(!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif))))
- Adds a prefetch(head->chain.first) in
__inet_lookup_established()/__tcp_v4_check_established() and
__inet6_lookup_established()/__tcp_v6_check_established() and
__dccp_v4_check_established() to bring into cache the first element of the
list, before the {read|write}_lock(&head->lock);
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Acked-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
I've found the problem in general. It affects any 64-bit
architecture. The problem occurs when you change the system time.
Suppose that when you boot your system clock is forward by a day.
This gets recorded down in skb_tv_base. You then wind the clock back
by a day. From that point onwards the offset will be negative which
essentially overflows the 32-bit variables they're stored in.
In fact, why don't we just store the real time stamp in those 32-bit
variables? After all, we're not going to overflow for quite a while
yet.
When we do overflow, we'll need a better solution of course.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
From: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Handle better the case where the sender sends full sized
frames initially, then moves to a mode where it trickles
out small amounts of data at a time.
This known problem is even mentioned in the comments
above tcp_grow_window() in tcp_input.c, specifically:
...
* The scheme does not work when sender sends good segments opening
* window and then starts to feed us spagetti. But it should work
* in common situations. Otherwise, we have to rely on queue collapsing.
...
When the sender gives full sized frames, the "struct sk_buff" overhead
from each packet is small. So we'll advertize a larger window.
If the sender moves to a mode where small segments are sent, this
ratio becomes tilted to the other extreme and we start overrunning
the socket buffer space.
tcp_clamp_window() tries to address this, but it's clamping of
tp->window_clamp is a wee bit too aggressive for this particular case.
Fix confirmed by Ion Badulescu.
Signed-off-by: David S. Miller <davem@davemloft.net>
But retain the comment fix.
Alexey Kuznetsov has explained the situation as follows:
--------------------
I think the fix is incorrect. Look, the RFC function init_cwnd(mss) is
not continuous: f.e. for mss=1095 it needs initial window 1095*4, but
for mss=1096 it is 1096*3. We do not know exactly what mss sender used
for calculations. If we advertised 1096 (and calculate initial window
3*1096), the sender could limit it to some value < 1096 and then it
will need window his_mss*4 > 3*1096 to send initial burst.
See?
So, the honest function for inital rcv_wnd derived from
tcp_init_cwnd() is:
init_rcv_wnd(mss)=
min { init_cwnd(mss1)*mss1 for mss1 <= mss }
It is something sort of:
if (mss < 1096)
return mss*4;
if (mss < 1096*2)
return 1096*4;
return mss*2;
(I just scrablled a graph of piece of paper, it is difficult to see or
to explain without this)
I selected it differently giving more window than it is strictly
required. Initial receive window must be large enough to allow sender
following to the rfc (or just setting initial cwnd to 2) to send
initial burst. But besides that it is arbitrary, so I decided to give
slack space of one segment.
Actually, the logic was:
If mss is low/normal (<=ethernet), set window to receive more than
initial burst allowed by rfc under the worst conditions
i.e. mss*4. This gives slack space of 1 segment for ethernet frames.
For msses slighlty more than ethernet frame, take 3. Try to give slack
space of 1 frame again.
If mss is huge, force 2*mss. No slack space.
Value 1460*3 is really confusing. Minimal one is 1096*2, but besides
that it is an arbitrary value. It was meant to be ~4096. 1460*3 is
just the magic number from RFC, 1460*3 = 1095*4 is the magic :-), so
that I guess hands typed this themselves.
--------------------
Signed-off-by: David S. Miller <davem@davemloft.net>
When you've enabled conntrack and NAT as a module (standard case in all
distributions), and you've also enabled the new conntrack netlink
interface, loading ip_conntrack_netlink.ko will auto-load iptable_nat.ko.
This causes a huge performance penalty, since for every packet you iterate
the nat code, even if you don't want it.
This patch splits iptable_nat.ko into the NAT core (ip_nat.ko) and the
iptables frontend (iptable_nat.ko). Threfore, ip_conntrack_netlink.ko will
only pull ip_nat.ko, but not the frontend. ip_nat.ko will "only" allocate
some resources, but not affect runtime performance.
This separation is also a nice step in anticipation of new packet filters
(nf-hipac, ipset, pkttables) being able to use the NAT core.
Signed-off-by: Harald Welte <laforge@netfilter.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
The GRE, SCTP and TCP protocol helpers did not call
ip_conntrack_event_cache() when updating ct->status. This patch adds
the respective calls.
Signed-off-by: Harald Welte <laforge@netfilter.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
We have to introduce a separate Kconfig menu entry for the NFQUEUE targets.
They cannot "just" depend on nfnetlink_queue, since nfnetlink_queue could
be linked into the kernel, whereas iptables can be a module.
Signed-off-by: Harald Welte <laforge@netfilter.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch fixes a number of bugs. It cannot be reasonably split up in
multiple fixes, since all bugs interact with each other and affect the same
function:
Bug #1:
The event cache code cannot be called while a lock is held. Therefore, the
call to ip_conntrack_event_cache() within ip_ct_refresh_acct() needs to be
moved outside of the locked section. This fixes a number of 2.6.14-rcX
oops and deadlock reports.
Bug #2:
We used to call ct_add_counters() for unconfirmed connections without
holding a lock. Since the add operations are not atomic, we could race
with another CPU.
Bug #3:
ip_ct_refresh_acct() lost REFRESH events in some cases where refresh
(and the corresponding event) are desired, but no accounting shall be
performed. Both, evenst and accounting implicitly depended on the skb
parameter bein non-null. We now re-introduce a non-accounting
"ip_ct_refresh()" variant to explicitly state the desired behaviour.
Signed-off-by: Harald Welte <laforge@netfilter.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Harald Welte <laforge@netfilter.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
As noted by Alexey Dobriyan, the DEBUGP statement prints the wrong
callID.
Signed-off-by: Harald Welte <laforge@netfilter.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since the introduction of TSO pcount a year ago, it has been possible
for tcp_fragment() to cause packets_out to decrease. Prior to that,
tcp_retrans_try_collapse() was the only way for that to happen on the
retransmission path.
When this happens with Reno, it is possible for sasked_out to become
invalid because it is only an estimate and not tied to any particular
packet on the retransmission queue.
Therefore we need to adjust sacked_out as well as left_out in the Reno
case. The following patch does exactly that.
This bug is pretty difficult to trigger in practice though since you
need a SACKless peer with a retransmission that occurs just as the
cached MTU value expires.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Patch from Joel Sing to fix the default congestion control algorithm
for incoming connections. If a new congestion control handler is added
(via module), it should become the default for new
connections. Instead, the incoming connections use reno. The cause is
incorrect initialisation causes the tcp_init_congestion_control()
function to return after the initial if test fails.
Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Acked-by: Ian McDonald <imcdnzl@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Cleanup the printk's in fib_trie:
* Convert a couple of places in the dump code to BUG_ON
* Put log level's on each message
The version message really needed the message since it leaks out
on the pretty Fedora bootup.
Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Acked-by: Robert Olsson <Robert.Olsson@data.slu.se>,
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix unchecked __get_user that could be tricked into generating a
memory read on an arbitrary address. The result of the read is not
returned directly but you may be able to divine some information about
it, or use the read to cause a crash on some architectures by reading
hardware state. CAN-2004-2492.
Fix from Al Viro, ack from Dave Miller.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The problem is that we're now calling tcp_fragment() in a context
where the packets might be marked as SACKED_ACKED or SACKED_RETRANS.
This was not possible before as you never retransmitted packets that
are so marked.
Because of this, we need to adjust sacked_out and retrans_out in
tcp_fragment(). This is exactly what the following patch does.
We also need to preserve the SACKED_ACKED/SACKED_RETRANS marking
if they exist.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Those exports are needed by the PPTP helper following in the next
couple of changes.
Signed-off-by: Harald Welte <laforge@netfilter.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Both __ip_conntrack_expect_find and ip_conntrack_expect_find_get take
a reference to the expectation, the difference is that callers of
__ip_conntrack_expect_find must hold ip_conntrack_lock.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Harald Welte <laforge@netfilter.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
This new "version 3" PPTP conntrack/nat helper is finally ready for
mainline inclusion. Special thanks to lots of last-minute bugfixing
by Patric McHardy.
Signed-off-by: Harald Welte <laforge@netfilter.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
* This patch is from Paul McKenney's RCU reviewing.
Signed-off-by: Robert Olsson <robert.olsson@its.uu.se>
Signed-off-by: David S. Miller <davem@davemloft.net>
* Prints the route tnode and set the stats level deepth as before.
Signed-off-by: Robert Olsson <robert.olsson@its.uu.se>
Signed-off-by: David S. Miller <davem@davemloft.net>
ip_ct_refresh_acct() can be called without a valid "skb" pointer.
This used to work, since ct_add_counters() deals with that fact.
However, the recently-added event cache doesn't handle this at all.
This patch is a quick fix that is supposed to be replaced soon by a cleaner
solution during the pending redesign of the event cache.
Signed-off-by: Harald Welte <laforge@netfilter.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Instead of maintaining an array containing a list of nodes this instance
is responsible for let's use a simple bitmap. This provides the
following features:
* clusterip_responsible() and the add_node()/delete_node() operations
become very simple and don't need locking
* the config structure is much smaller
In spite of the completely different internal data representation the
user-space interface remains almost unchanged; the only difference is
that the proc file does not list nodes in the order they were added.
(The target info structure remains the same.)
Signed-off-by: KOVACS Krisztian <hidden@balabit.hu>
Signed-off-by: Harald Welte <laforge@netfilter.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
The CLUSTERIP target creates a procfs entry for all different cluster
IPs. Although more than one rules can refer to a single cluster IP (and
thus a single config structure), removal of the procfs entry is done
unconditionally in destroy(). In more complicated situations involving
deferred dereferencing of the config structure by procfs and creating a
new rule with the same cluster IP it's also possible that no entry will
be created for the new rule.
This patch fixes the problem by counting the number of entries
referencing a given config structure and moving the config list
manipulation and procfs entry deletion parts to the
clusterip_config_entry_put() function.
Signed-off-by: KOVACS Krisztian <hidden@balabit.hu>
Signed-off-by: Harald Welte <laforge@netfilter.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
ip_vs_ftp when loaded can create NAT connections with unknown client
port for passive FTP. For such expectations we lookup with cport=0 on
incoming packet but it matches the format of the persistence templates
causing packets to other persistent virtual servers to be forwarded to
real server without creating connection. Later the reply packets are
treated as foreign and not SNAT-ed.
This patch changes the connection lookup for packets from clients:
* introduce IP_VS_CONN_F_TEMPLATE connection flag to mark the
connection as template
* create new connection lookup function just for templates -
ip_vs_ct_in_get
* make sure ip_vs_conn_in_get hits only connections with
IP_VS_CONN_F_NO_CPORT flag set when s_port is 0. By this way
we avoid returning template when looking for cport=0 (ftp)
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: David S. Miller <davem@davemloft.net>
Agostino di Salle noticed that persistent templates are not
invalidated due to buggy optimization.
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch fixes line dupes at /ipv4/igmp.c and /ipv6/mcast.c in the
2.6 kernel, where MCAST_EXCLUDE is mistakenly used instead of
MCAST_INCLUDE.
Signed-off-by: Denis Lukianov <denis@voxelsoft.com>
Signed-off-by: David L Stevens <dlstevens@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The problem is that the SACK fragmenting code may incorrectly call
tcp_fragment() with a length larger than the skb->len. This happens
when the skb on the transmit queue completely falls to the LHS of the
SACK.
And add a BUG() check to tcp_fragment() so we can spot this kind of
error more quickly in the future.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
In 2.6.13-rcX the MASQUERADE target was changed not to exclude local
packets for better source address consistency. This breaks DHCP clients
using UDP sockets when the DHCP requests are caught by a MASQUERADE rule
because the MASQUERADE target drops packets when no address is configured
on the outgoing interface. This patch makes it ignore packets with a
source address of 0.
Thanks to Rusty for this suggestion.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Don't parse the packet, the data is already available in the conntrack
structure.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
With large port numbers the helper_names buffer can overflow.
Noticed by Samir Bellabes <sbellabes@mandriva.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use schedule_timeout_{,un}interruptible() instead of
set_current_state()/schedule_timeout() to reduce kernel size. Also use
human-time conversion functions instead of hard-coded division to avoid
rounding issues.
Signed-off-by: Nishanth Aravamudan <nacc@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
There is an extra left_out/lost_out adjustment in tcp_fragment which
means that the lost_out accounting is always wrong. This patch removes
that chunk of code.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Clean up timer initialization by introducing DEFINE_TIMER a'la
DEFINE_SPINLOCK. Build and boot-tested on x86. A similar patch has been
been in the -RT tree for some time.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
With the use of RCU in files structure, the look-up of files using fds can now
be lock-free. The lookup is protected by rcu_read_lock()/rcu_read_unlock().
This patch changes the readers to use lock-free lookup.
Signed-off-by: Maneesh Soni <maneesh@in.ibm.com>
Signed-off-by: Ravikiran Thirumalai <kiran_th@gmail.com>
Signed-off-by: Dipankar Sarma <dipankar@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Create one iterator for walking over FIB trie, and use it
for all the /proc functions. Add a /proc/net/route
output for backwards compatibility with old applications.
Make initialization of fib_trie same as fib_hash so no #ifdef
is needed in af_inet.c
Fixes: http://bugzilla.kernel.org/show_bug.cgi?id=5209
Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
One such place that can damage the dst refcnts is route.c with
CONFIG_IP_ROUTE_MULTIPATH_CACHED enabled, i don't see the user's
.config. In this new code i see that rt_intern_hash is called before
dst->refcnt is set to 1, dst is the 2nd arg to rt_intern_hash.
Arg 2 of rt_intern_hash must come with refcnt 1 as it is added to
table or dropped depending on error/add/update. One such example is
ip_mkroute_input where __mkroute_input return rth with refcnt 0 which
is provided to rt_intern_hash. ip_mkroute_output looks like a 2nd such
place. Appending untested patch for comments and review. The idea is
to put previous reference as we are going to return next result/error.
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: David S. Miller <davem@davemloft.net>
A UDP packet may contain extra data that needs to be trimmed off.
But when doing so, UDP forgets to fixup the skb checksum if CHECKSUM_HW
is being used.
I think this explains the case of a NFS receive using skge driver
causing 'udp hw checksum failures' when interacting with a crufty
settop box.
Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
This was found by inspection while looking for checksum problems
with the skge driver that sets CHECKSUM_HW. It did not fix the
problem, but it looks like it is needed.
If IP reassembly is trimming an overlapping fragment, it
should reset (or adjust) the hardware checksum flag on the skb.
Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
The following patch kills __ip_ct_expect_unlink_destroy and export
unlink_expect as ip_ct_unlink_expect. As it was discussed [1], the function
__ip_ct_expect_unlink_destroy is a bit confusing so better do the following
sequence: ip_ct_destroy_expect and ip_conntrack_expect_put.
[1] https://lists.netfilter.org/pipermail/netfilter-devel/2005-August/020794.html
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller